University of Oxford Crest

Laboratory of Molecular Biophysics
Laboratory Journal 2002
Declan A. Doyle


Return to: Contents.

Declan A. Doyle


Ion channels and transporters



Declan A. Doyle, Jonathan Cuthbertson, Anling Kuo, Tahmina Rahman, Jochen Zimmer

    Two major hurdles must be overcome if the structure of a membrane protein is to be determined. The first involves obtaining protein in sufficient quantities to be able to screen for suitable crystals.  Membrane proteins that are naturally abundant are good targets such as those involved in the respiratory chain or photosynthesis.  Alternatively, the target protein can be over-expressed in expression systems ranging from bacteria to mammalian cells. To date, only the Escherichia coli expression system has been successful in providing membrane protein in sufficient quantities and in a homogeneous state that has lead to structural determination (Chang et al., 1998; Doyle et al., 1998; Dutzler et al., 2002). The next problem is finding a crystal that is sufficiently well ordered and diffracts to a limit that would make the project feasible. For membrane proteins this can be a real problem because of the flexible, dynamic detergent micelle that surrounds the protein. The detergent micelle is necessary for the maintenance of structural integrity of the membrane protein outside of a phospholipid bilayer. Generally, for protein crystallography, large degrees of flexibility within the protein are considered detrimental for high resolution crystals. To overcome this problem with soluble proteins alternative constructs that explore domains within the structure or homologous proteins can be considered. Repeating this method with membrane proteins does not improve the situation as in all cases the detergent will still be present.
    A number of methods have been developed to overcome the detergent problem in membrane protein structure determination. Increasing the soluble protein surface area with the use of antibody fragments is one approach. This method is believed to improve the quality of the crystal by increasing the number of protein-protein contacts within the lattice and decreasing the weaker and less stable protein-detergent or detergent-detergent contacts. In the bicontinous cubic phase method the detergent is no longer required during the crystallisation step as the protein is re-introduced into a membrane bilayer environment (Landau and Rosenbusch, 1996).
    During our structural studies of the E. coli membrane protein yadQ we uncovered a novel method for improving the diffraction limits and integrity of a membrane protein crystal (Kuo, et al, 2002). The improvement was brought about by the simple process of dehydration. Slow evaporation of water over a period of months resulted in the improvement in the resolution limits from 8.0 to 4.0 Å with a concurrent decrease in the mosaicity to values of approximately 1° (Figure 1).
YadQ is a member of the voltage-dependent chloride channel family. The channel, as for all others within this family, forms homodimers as the active complex within a membrane.  It has been shown functionally for the eukaryotic homologue of the bacterial chloride channel that each monomer has a separate ion conduction pathway that opens and closes independently (Middleton et al., 1996). This mechanism is referred to as the fast gate. Another gating mechanism shuts both monomers simultaneously, referred to as the slow gate. It is conceivable that a conformation change occurs at the dimer interface during slow gating. We can only speculate as to the exact conformation change that the channel undergoes upon gating but such a change may account for the variation in unit cell dimensions that we observe upon dehydration. Equally as likely is the possibility that the molecule is relatively rigid and the dehydration process causes the channel to go through a rigid-body rotation. This may be aided by the presence of the flexible detergent micelle. In this scenario, the micelle changes its structure during the dehydration process allowing chloride channel to rotate.

Heavy metal ATPases use the hydrolysis of ATP to move specific ions against their concentration gradient. It is believed that all of the members within the ATPase family follow a similar reaction scheme to that of the calcium-ATPase. This involves binding of ions and ATP followed by phosphorylation of the enzyme. The ATPase then undergoes a conformational change resulting in the release of the ions. Dephosphorylation of the enzyme completes the cycle and the protein sets up to once again accept ions and ATP. Many detailed studies have characterised the various steps in this cycle and have demonstrated that for optimum activity the enzyme requires particular closely associated lipids. We have begun to characterise the lipid requirements of the zinc-ATPase. This protein has a central core that is structurally related to the calcium ATPase based on sequence alignments.  Initial studies have indicated that, like the calcium ATPase, the zinc enzyme's activity is enhanced in the presence of specific lipids. Further studies will be aimed at uncovering the specific lipid nature of the enzyme activity and domain movement as the protein goes through the catalytic cycle.   
It is believed that the distinct environment that a membrane protein resides in dictates the type of structure that is formed within the bilayer region, i.e. only transmembrane alpha helices or only beta sheet. Several sophisticated transmembrane helical prediction programs are publicly available that identify potential regions within the linear amino acid sequence that span the membrane. At present we are analysing these predictions against the known membrane structure database. This information should provide a clearer picture of the ability to correctly predict transmembrane helical segments.  It is hoped that with this knowledge experimentalists can fragment the target protein into it's structural components with a greater degree of confidence.

Diffraction patterns and morphologies of crystals of the integral membrane protein yadQ
Figure 1. Diffraction patterns and morphologies of crystals of the integral membrane protein yadQ ...more



References

Chang, G., Spencer, R. H., Lee, A. T., Barclay, M. T., Rees, D. C. (1998) Structure of the MscL homolog from Mycobacterium tuberculosis: a gated mechanosensitive ion channel. Science. 282, 2220-2226.
Doyle, D. A., Morais-Cabral, J., Pfuetzner, R. A., Kuo, A., Gulbis, J. M., Cohen, S. L., Chait, B. T., MacKinnon, R. (1998) The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science. 280, 69-77.
Dutzler, R., Campbell, E. B., Cadene, M., Chait, B. T., MacKinnon, R. (2002) X-ray structure of a ClC chloride channel at 3.0 Å reveals the molecular basis of anion selectivity. Nature. 415, 287-294.
Kuo, A., Bowler, M. W., Zimmer, J., Antcliff, J. F., Doyle, D. A. (2002) Increasing the diffraction limits and internal order of a membrane protein crystal by dehydration.  J. Struct. Biol. In press.
Landau, E. M., Rosenbusch, J. P. (1996) Lipidic cubic phases: a novel concept for the crystallization of membrane proteins. Proc. Natl. Acad. Sci. 93, 14532-14535.
Middleton, R. E., Pheasant, D. J., Miller, C. (1996) Homodimeric architecture of a ClC-type chloride ion channel. Nature. 383, 337-340.




Return to: Contents.


[ Lab. of Molecular Biophysics | University of Oxford | Feedback ]

Last updated: 16-MAY-2003 13:18